Introduction

This document is part instruction manual and part data summary tool focused on the collection, wranggling, and summarizing passive audio recordings (PAR) data. Data are generated by automated recording units (ARUs) deployed along an urban rural gradient from Newark, DE to Landenberg, PA.

The first step in this process is to deploy the ARU’s at the appropriate sites (See map and table below) with fresh batteries and formatted SD Cards. We have found that the AudioMoth ARUs will last 1.5 months when they have fresh AA batteries and are recording continuously (10 minute file saves) for two hours in the morning and two hours in the evening.

Carolina Wren Vocalization types

We selected Carolina Wren song vocalizations to train the BirdNET model and screen the XX months of recordings. Carolina Wrens vocalization types used in this analysis include;

Figure 1. Male Carolina Wren Song A
Figure 1. Male Carolina Wren Song A



Figure 2. Male Carolina Wren Song B
Figure 2. Male Carolina Wren Song B



Figure 3. Male Carolina Wren Song Bout
Figure 3. Male Carolina Wren Song Bout



Figure 4. Male Carolina Wren Song

Methods

We located 20 Automated Recording Units (ARUs) along an urban / rural gradient, like so many others have done before us! We placed ARUs at locations where Carolina Wrens were detected on at least two of three visits with the motivation that our primary objective was to describe the phenological variation in Carolina Wren vocalizations over the annual cycle.

Figure 5. Study sites
Figure 5. Study sites



Table 1. Study site ID, Lat, Long, # hours recorded

Site_ID latitude longitude
1 39.76574 -75.78674
2 39.75722 -75.78384
3 39.75202 -75.77538
4 39.74589 -75.77399
5 39.73643 -75.76424
6 39.71957 -75.76209
7 39.71193 -75.75936
8 39.71631 -75.76090
9 39.72483 -75.76762
10 39.72924 -75.76563
11 39.68237 -75.74138
12 39.69333 -75.75937
13 39.69598 -75.75568
14 39.68959 -75.75491
15 39.68782 -75.74917
16 39.69389 -75.76287
17 39.66918 -75.75050
18 39.67336 -75.74437
19 39.67873 -75.74261
20 39.67142 -75.74995

Carolina Wren Vocalization Phenology along an urban / rural gradient. Sites 01 - 10 are ‘rural’ and Sites 11 - 20 are ’urban.



## ADD lapply code for making am pm figures here...


song.dat <- read.csv('../Reality-Wren/Output/wren_detections.csv')

## Create a column for morning (am) and evening (pm)

song.dat <- song.dat %>%
  mutate(time_code = case_when(time < 1200 ~ 'am',
                               time > 1200 ~ 'pm'))


## Create df grouped by time_code (am & pm)

total_wren_day <- song.dat %>% 
  group_by(site, month, day, time_code) %>%
  summarize(total_wren = sum(count))


total_wren_day$site <- as.factor(total_wren_day$site)

sites <- levels(total_wren_day$site)

## create a plot function that will use lapply to iterate through sites making a figure for each site (like above but this time with stacked bars showing songs in am and pm)


## Get file names to change
my_plot_hours <- function(sites){
  total_wren_day %>%
    filter(site == sites) %>%
    ggplot(aes(fill=time_code, y=total_wren, x=day)) + 
    geom_bar(position="stack", stat="identity")+
    theme_classic() +
    labs(
      x = "day",
      y = "total wren detections",
      title = paste("Reality Wren Phenology Mar-Oct Site", sites)) +
    facet_grid(rows=vars(month))
  }

lapply(sites, my_plot_hours)
## [[1]]

## 
## [[2]]

## 
## [[3]]

## 
## [[4]]

## 
## [[5]]

## 
## [[6]]

## 
## [[7]]

## 
## [[8]]

## 
## [[9]]

## 
## [[10]]

## 
## [[11]]

## 
## [[12]]

## 
## [[13]]

## 
## [[14]]

## 
## [[15]]

## 
## [[16]]

## 
## [[17]]

## 
## [[18]]

## 
## [[19]]

## 
## [[20]]

## run the vector of sites through the function using lapply
#htmltools::tagList(lapply(sites, my_plot_hours))